Modeling chloride penetration in self-consolidating concrete using artificial neural network combined with artificial bee colony algorithm

Abstract

This paper examines the robustness of a feed forward artificial neural network combined with an artificial bee colony algorithm (FF-ABC) in the prediction of chloride penetration in self-consolidating concretes. To this end, several self-consolidating concrete mixes were made using various mix proportions, and their rapid chloride penetrations (RCPT) were measured. The mix proportions and RCPT results were used as input and output variables, respectively, to train and test the proposed method. To verify accuracy of the FF-ABC model, its performance was compared to linear regression, genetic algorithm (GA), and particle swarm optimization (PSO) models. This comparison was conducted in three stages of training, validation, and testing. Results of this study indicate higher reliability of the FF-ABC model in comparison with the statistical, GA, and PSO models.